: AI Architect – Generative AI & Agentic AI

Experience: 7–8 Years

Industry: Banking / Financial Services preferred

Location: Mumbai(Thane /BKC) Work from Office

Role Type: Architecture, Engineering & Technology Leadership

1. About the Role

We are seeking a highly skilled AI Architect with 7–8 years of experience in designing, developing, and implementing enterprise-grade AI solutions. The ideal candidate will combine strong software engineering fundamentals with deep expertise in Generative AI, Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), AI agents, and cloud-native architectures. The candidate will be responsible for translating business requirements into scalable, secure, resilient, and production-ready AI solutions, with a focus on banking and financial services use cases. The role requires hands-on technical leadership, architectural decision-making, rapid prototyping, and collaboration with engineering, data, security, and business teams.
1. Key Responsibilities

AI Architecture & Solution Design
- Define end-to-end architecture for enterprise AI, GenAI, and agentic AI solutions aligned with business objectives and enterprise architecture standards.
- Design scalable AI platforms integrating LLMs, enterprise data sources, APIs, microservices, and cloud-native infrastructure.
- Evaluate and select appropriate foundation models, embedding models, orchestration frameworks, vector databases, and inference strategies.
- Establish reusable AI architecture patterns, reference implementations, design principles, and engineering standards.
- Design for high availability, scalability, low latency, cost optimisation, observability, and resilience.

Generative AI & Agentic AI Engineering
- Architect and implement RAG pipelines, including document ingestion, chunking, embedding generation, retrieval, reranking, and contextual response generation.
- Design intelligent AI agents capable of tool calling, multi-step reasoning, workflow orchestration, and controlled execution of enterprise business processes.
- Develop multi-agent workflows where appropriate, with clear boundaries, state management, human approval, and failure-handling mechanisms.
- Implement prompt engineering, structured outputs, model routing, context management, and conversation memory.
- Apply guardrails to mitigate hallucinations, prompt injection, sensitive-data exposure, and unauthorised tool execution.
- Build proof of concepts and convert successful prototypes into production-ready solutions.

Enterprise Integration & Banking Use Cases
- Integrate AI capabilities with enterprise applications, REST APIs, event-driven systems, microservices, and secure data platforms.
- Design AI solutions for banking use cases such as customer service automation, knowledge assistants, document intelligence, regulatory compliance, fraud investigation support, and developer productivity.
- Ensure AI services integrate seamlessly with existing authentication, authorisation, audit, and monitoring frameworks.
- Work with domain architects and business stakeholders to identify high-value AI opportunities and translate them into technical solutions.

MLOps, LLMOps & Production Engineering




- Establish CI/CD pipelines and automated deployment practices for AI applications and model-serving components.
- Implement model evaluation, regression testing, prompt/version management, retrieval-quality assessment, and continuous monitoring.
- Define evaluation frameworks covering factual accuracy, groundedness, relevance, latency, safety, and token consumption.
- Monitor model performance, drift, failure rates, inference costs, and application reliability.
- Optimise inference performance through caching, batching, model selection, quantisation, and other suitable techniques.

Security, Governance & Responsible AI
- Design AI solutions with security and privacy by design, including encryption, role-based access control, secrets management, and data minimisation.
- Implement responsible AI controls, explainability where appropriate, traceability, audit logging, and human-in-the-loop approvals.
- Address threats such as prompt injection, data poisoning, insecure output handling, excessive agent permissions, and sensitive information leakage.
- Collaborate with information security, risk, compliance, and data governance teams to meet applicable regulatory and organisational requirements.

Technical Leadership & Collaboration
- Provide hands-on technical leadership to developers and engineering teams throughout the solution lifecycle.
- Conduct architecture reviews, code reviews, technical design sessions, and proof-of-concept evaluations.
- Mentor engineers on AI engineering best practices, design patterns, and production-readiness standards.
- Communicate architecture decisions, trade-offs, risks, and implementation roadmaps to technical and business stakeholders.

1. Required Technical Skills

Generative AI & LLMs
- Strong understanding of LLMs, transformer-based models, embeddings, tokenisation, context windows, and inference.
- Hands-on experience with commercial and/or open-source models such as GPT, Claude, Gemini, Llama, or equivalent.
- Expertise in RAG, semantic search, hybrid retrieval, reranking, prompt engineering, and structured generation.
- Practical experience with agent frameworks such as LangGraph, LangChain, Semantic Kernel, or equivalent.

Programming & Backend Engineering
- Strong Python programming skills and experience building production-grade AI applications.
- Solid understanding of REST APIs, asynchronous processing, API security, and distributed systems.
- Experience with FastAPI, Flask, or equivalent backend frameworks.
- Working knowledge of Java/Spring Boot and microservices is an advantage.

Data & Search Technologies
- Experience with vector databases or vector search technologies such as pgvector, Azure AI Search, Pinecone, Weaviate, or Milvus.
- Strong understanding of SQL/NoSQL databases, document processing,



data pipelines, and data quality.
- Familiarity with knowledge graphs, hybrid search, and enterprise data integration is desirable.

Cloud & Platform Engineering
- Hands-on experience with at least one major cloud platform: Microsoft Azure, AWS, or Google Cloud.
- Experience deploying AI workloads using Docker, Kubernetes, and container-based architectures.
- Familiarity with API gateways, event-driven architecture, secrets management, monitoring, and distributed tracing.
- Experience with cloud AI platforms such as Azure AI Foundry, Amazon Bedrock, Vertex AI, or equivalent.

AI Quality, Evaluation & Operations
- Experience with LLM evaluation frameworks, automated testing, tracing, and observability tools.
- Understanding of hallucination mitigation, grounding, model safety, and retrieval-quality optimisation.
- Knowledge of CI/CD, infrastructure as code, and MLOps/LLMOps practices.

1. Qualifications & Experience

* 7–8 years of overall experience in software engineering, AI/ML engineering, solution architecture, or related technology roles.
- Demonstrable hands-on experience delivering AI or GenAI solutions beyond proof-of-concept stage.
- Robust architectural problem-solving skills and ability to make informed build-versus-buy and model-selection decisions.
- Experience working with enterprise-scale applications, sensitive data, and production deployment constraints.
- Bachelor’s or master’s degree in Computer Science, Information Technology, Engineering, or a related field.
- Banking, financial services, insurance, or other regulated-industry experience is highly desirable.

1. Preferred Experience

* Experience building agentic AI systems that invoke enterprise tools and APIs under controlled permissions.
- Knowledge of AI security testing, red teaming, model risk management, and responsible AI practices.
- Experience with multimodal AI, document intelligence, speech, or vision-based solutions.
- Familiarity with Kafka or other event-streaming platforms and integration with microservices.
- Experience designing AI solutions for high-volume, low-latency, business-critical environments.

1. Success Criteria

The AI Architect will be measured on the ability to:
- Deliver secure, scalable, production-ready AI solutions aligned with business outcomes.
- Establish reusable AI architecture patterns that accelerate delivery across engineering teams.
- Improve answer quality, retrieval accuracy, reliability, and user adoption.
- Optimise inference cost, response latency, and infrastructure utilisation.
- Demonstrate measurable business value through automation, productivity improvements, and reduced manual effort.
- Ensure AI solutions meet defined security, governance, auditability, and operational-readiness requirements.

Ideal Candidate Profile: A hands-on AI architect who can move from business problem definition to solution architecture, working code, production deployment, and measurable business impact—not just develop GenAI proofs of concept.

Pay: ₹1,500,000.00 - ₹2,500,000.00 per year

Experience

- Agentic AI & gen AI: 5 years (Required)

Work Location: In person

📌 AI Architect(Gen AI & Agentic AI) 7 to 8yrs (Mumbai)
🏢 Skill-mine Technology
📍 Mumbai